Early albumin administration and mortality in ICU patients with heart failure and hypoalbuminemia: a target trial emulation

Abstract

Abstract Background Whether albumin improves outcomes in critically ill patients with heart failure and hypoalbuminemia remains uncertain. We aimed to estimate the per-protocol effect of albumin initiation after a 6-h landmark on 28-day mortality in this population. Methods We emulated a parallel-group target trial using the Medical Information Mart for Intensive Care IV (MIMIC-IV; version 3.1; 2008–2022). Adults with a first intensive care unit (ICU) admission, heart failure identified using International Classification of Diseases (ICD) codes, and serum albumin < 3.0 g/dL were eligible if they remained in the ICU and had not received albumin by 6 h after admission. We compared albumin initiation from 6 to < 48 h after ICU admission with no albumin during this window. The primary outcome was 28-day all-cause mortality. Per-protocol risk differences (RDs) and hazard ratios (HRs) were estimated using cloning, artificial censoring, and stabilized inverse probability of censoring weighting. Results Among 1,308 patients, 80 initiated albumin and 1,228 did not. Weighted 28-day cumulative mortality was 32.9% and 30.7%, respectively (RD, 2.2%; 95% confidence interval [CI], − 7.6% to 12.0%; HR, 1.037; 95% CI, 0.890 to 1.207). Estimates at 90 days and 1 year also had confidence intervals spanning the null. Sensitivity analyses were broadly consistent with the primary analysis, while subgroup estimates were imprecise. Conclusions Albumin initiation from 6 to < 48 h after ICU admission did not demonstrate lower 28-day mortality. The small treated group and wide confidence intervals do not exclude clinically meaningful benefit or harm. These findings do not address albumin use during the first 6 h of ICU care.